An Empirical Study on the Artificial Intelligence Writing Evaluation System in China CET

Area of Science:

• Educational Technology
• Natural Language Processing
• Applied Linguistics

Background:

• Automated Writing Evaluation (AWE) systems are increasingly utilized in Chinese higher education for college English writing.
• These systems aim to reduce teacher workload and enhance student writing proficiency.
• Juku AWE is a prominent system adopted by numerous Chinese colleges and universities.

Purpose of the Study:

• To empirically investigate the effectiveness and reception of the Juku Automated Writing Evaluation (AWE) system in college English writing instruction.
• To assess student writing improvement and gather feedback from both students and teachers on AWE usage.

Main Methods:

• An empirical study involving 114 students across 2 classes at Xi'an University.
• Data collection through questionnaires and interviews with 30 teachers and 200 students using Juku AWE.

Main Results:

• The study confirmed that AWE systems significantly aid students in improving their English writing abilities.
• Both teachers and students reported positive attitudes towards AWE, citing benefits like immediate feedback, time savings, and increased engagement.
• Limitations were identified, with AWE systems needing improvement in evaluating text structure, logical coherence, and content.

Conclusions:

• Automated Writing Evaluation (AWE) offers tangible benefits for college English writing, enhancing both learning and teaching efficiency.
• Despite its advantages, users must maintain a balanced perspective, recognizing the current limitations in comprehensive text analysis.
• Further development of AWE is recommended to address areas such as structural and logical evaluation for more holistic writing assessment.